Project Topics
Get Complete Project Material File(s) Now! »
VOC EMISSION RATIOS
The formation of methane is often accompanied by a number of other compounds, whose abundances depend strongly on the creation conditions. Thus, it is possible to use correlations of co-emitted compounds with methane to distinguish between individual sources.
The term volatile organic compounds (VOCs) is used to denote the entire set of vapour phase atmospheric organics excluding CO and CO2. Given their short atmospheric lifetimes (fractions of a day to weeks) they have little direct impact on radiative forcing but are central to atmospheric chemistry, participating in atmospheric photochemical reactions and influencing the air quality and climate through their production of ozone and organic aerosols. Within this thesis, non-methane hydrocarbons (NMHC), which are organic chemical compounds consisting of hydrogen and carbon atoms emitted from both natural and anthropogenic sources are used as complementary tracers for CH4. Here, the words NMCHs and VOCs are used interchangeably.
The majority of VOC emissions are related to natural sources which originate from nearly exclusively (approximately 90%) vegetation [Guenther et al., 1995]. Nonetheless, global emissions of anthropogenic VOCs is approximately 186 Tg/year [EDGAR 2005], of which a number of sources are shared with methane. The ratio of methane to light VOCs is very high for biologically produced methane because the biochemical mechanism for methanogensis are very specific, whereas in thermogenic reactions substantial amounts of ethane and propane can also be produced. VOCs can be separated into a number of sub-categories which can be used as trace gasses to identify methane sources, namely: alkanes, alkenes, alkynes, aromatics and oxygenated VOCs (OVOCs). The main sources of alkane emissions, such as ethane and propane, are from exploitation and distribution of natural gas, petrochemical industries and biomass burning. Fossil fuels contain only small amounts of alkenes, thus such VOCs (e.g. ethene and propene) are emitted predominantly from vehicle exhaust (due to incomplete combustion), from biofuel combustion and biomass burning. Aromatics, such as benzene, toluene, xylenes (BTEX) are components in fossil fuels, and are predominantly emitted by vehicle exhaust from fuel evaporation and spillage. Distinction between sources can sometimes be difficult as source characteristics vary spatially and temporally. For example, exhaust contribution to VOC levels were found to vary depending on the time of day and day of the week by Rubin et al. [2006]. Furthermore, the composition of the exhaust was found to be dependent on the type of vehicle and fuel used [Verma and des Tombe, 2002, Schuetzle et al., 1994, Zhao et al., 2011].
The use of emissions ratios is a widely-used method for determining source composition and allows for the separation of sources. In literature, this method has been predominantly used to characterise NMHCs [So et al., 2004, Wang et al., 2010]. Nonetheless there has been a recent surge in publications using VOC emissions ratios to identify and distinguish between thermogenic (in particular oil and gas) methane emissions. [Koss et al, 2015, Warneke et al., 2014, Gilman et al, Petron et al., 2014]. Oil and gas sources can be identified using a number of VOC:CH4 correlations, the predominant being ethane, (C2H6) which is the secondary component in natural gas, as well as other light hydrocarbons C1-C5. An example of how the C2H6:CH4 ratio can be used to identify gas of differing origins can be seen in Figure 1.4 from Schoell [1983]. The plot indicates that thermogenic gasses formed during or directly after the formation of oil (green regions) are much richer in C2+ hydrocarbons than dry gasses formed later (pink regions). Biogenic methane trace gases can be slightly more complex to distinguish; Yuan et al. 2017 found ammonia and ethanol to be good tracers for animal & waste emissions and feed storage & handling emissions respectively. The major co-emitted VOCs for anthropogenic methane sources can be found in Table 1.1.
Table 1-1 An example of co-emitted VOCs that can be used as tracers to identify specific methane sources. The VOC:CH4 ratio can depend on many environmental factors e.g. temperature, location etc., and thus can vary for each individual source and in time.
STABLE ISOTOPES OF METHANE
Another consequence of methane formation is that different methane processes result in different isotopic ratios of carbon (13C/12C) and hydrogen (D/H). It has been demonstrated that these stable isotope ratios can be used to identify methane sources because the isotopic signatures of different sources and sinks are unique [Schoell, 1983]. Carbon isotopes are the most frequently measured isotope ratios in atmospheric CH4 and are an integral part of this work, thus in this thesis I will focus on δ13C-CH4, which is also commonly abbreviated as δ13CH4.
Most methane on earth is composed of one atom of 12C and four atoms of 1H, however found in small quantities methane isotopologues containing heaver isotopes of carbon, namely 13C and 14C are also present. The abundance of the heavier isotopologues differs slightly between the land surface and atmosphere due to isotopic fractionation when methane is produced or consumed. Heavier isotopes have lower reaction rates, so emitted methane contains a lower fraction of heavier isotopes than the reaction substrate, while methane sinks lead to enrichment of reservoir CH4 in atmospheric heavy isotopologues as the reaction consumes preferably 12CH4. Thus, the isotopologue abundances of emitted methane depends strongly on the isotopic abundances in the organic matter substrate, which is relatively constant in time. Following this we can use the isotopic ratio to attribute a characteristic isotopic signature to each source process; the isotopic signature is as D and expressed as the relative deviation against the Vienna Peedee belemnite (VPDB) reference material. Because the variations that occur are on the order of one part in a thousand or smaller they are expressed in permil (‰) or parts per thousand:
The average δ13CH4 values of the contemporary atmosphere range about -47.5‰ [e.g. Quay et al., 1999] with an annual cycle resulting from the spatio-temporal distribution of sinks and sources and atmospheric transport [Hein et al., 1997, Quay et al., 1999, Stevens & Engelkemeir, 1988].
Using the method described by Equation 1.1 methane sources can be characterised by source specific isotopic signatures as they reflect different methane production processes. For example, methanogenisis results in emissions that are highly depleted in 13C (δ13CH4 is in the range of -60‰), whereas methane derived from biomass burning retains the isotopic characteristic of the fuel and is generally highly enriched in 13C compared to background atmospheric methane (ranging from -27‰ to – 18‰ depending on C-3 or C-4 plants). The typical ranges of δ13C of methane sources can be found in Table 2. Figure 1.4, taken from Schoell (1983), demonstrates how combining information of the methane isotopic signature and the concentration of hydrocarbons can be used to identify the origins of natural gas and petroleum, thus providing a means to fingerprint such methane sources.
In this way, isotopic measurements can aid in partitioning and identifying sources when measuring site scale emissions thus improving emission inventories, and also provide an additional constraint on the large uncertainties in the present methane budget estimates because the net isotopic composition of methane emissions depends on the balance of these different sources.
Figure 1.4 Genetic characterisation of natural gases by compositional and isotopic variations taken from Schoell (1983). A) A schematic illustration of the formation of natural gas and petroleum in relation to the maturity of organic matter B) The relative concentration of C2+ hydrocarbons in gases in relation to 13C concentration in methane. Biogenic gas is represented in yellow and by the letter B. There are two stages of thermogenic gas, T in light green which forms during or directly after oil formation, and deep dry gasses TT formed after the principle stages of oil formation (formed by humic, TT(h) and from marine source rocks TT(m) in light and dark pink respectively.
The overall aim of this thesis was to test and further develop the current methods used for anthropogenic methane source identification of site scale measurements. The primary objective was to improve on present source apportionment techniques, whilst the second objective was to apply the developed methods to separate methane sources at industrial sites. Throughout the thesis, a number of measurement field campaigns were undertaken, targeting the major industries contributing to methane emissions, namely: natural gas compressor station, oil extraction, wastewater treatment plants, landfill, and agriculture. The dates, locations and species measured at these sites can be seen in Table 1.3. The aim of such measurement campaigns was to gain an insight into the characteristics of emissions from a variety of industries, and to test different measurement methods to determine the most useful instruments and techniques to measure and separate methane sources for specific sites.
INSTRUMENTATION
Predominantly two instruments were used regularly throughout this thesis; Cavity Ring Down Spectroscopy (CRDS) measuring CH4, CO2, H2O, C2H6, δ13CH4 and δ13CO2, and Gas Chromatographs (GC) measuring light (C2-C5) VOCs. CRDS uses a single frequency laser diode to measure specific gas-phase molecules which scatter and absorb light in the near infrared absorption spectrum. By measuring the height of absorption peaks the concentrations of specific species can be determined. The CRDS instrument used throughout this thesis is a G2201-i Picarro. The GCs used in this thesis are based on flame ionisation detectors (FIDs) which measure the concentrations of organic species in a gas stream by detecting the ions formed during the combustion of organic compounds in a hydrogen flame. A manual GC (Chrompack Variean 3400) was used for measurements of flask samples while for continuous, field measurements an automatic GC (Chromatotec) was used. The instruments used and technical developments of CRDS are described in Chapter 2 which is based on the published study Assan et al., (2017).
The improvement and evaluation of source apportionment techniques specifically for methane source identification is explored in Chapter 3. Three methods are applied to continuous measurements of CH4 and VOCs taken at a natural gas compressor station campaign, and compared, namely; carbon isotopes in methane, principle component analysis (PCA) and positive matrix factorisation (PMF). PCA and PMF are linear receptor models often used in PM studies to identify contributions of different sources to local concentration enhancements. Source profiles and contributions are calculated on the basis of correlations within the data, which assumes that highly correlated variables originate from the same source. Chapter 3 describes these models and details developments made to enhance their CH4 source identification potential. A sensitivity study of PCA and PMF can be found at the end of Chapter 3. All three methods are used to analyse data from a natural gas compressor station, and results compared.
APPLICATION TO FIELD MEASUREMENTS
In the final Chapter of this thesis, the source apportionment techniques developed are implemented on data taken from 6 measurement campaigns. The sites constitute the major biogenic CH4 sources in the Ile de France region: livestock, wastewater and landfill. Chapter 4 characterises these sources using isotopic analysis, source ratios of co-emitted species and, receptor models.
Chapter 2 CHARACTERISATION OF INTERFERENCES TO IN-SITU OBSERVATIONS OF METHANE ISOTOPES AND C2H6 WHEN USING A CAVITY RING DOWN SPECTROMETER AT INDUSTRIAL SITES.
SUMMARY OF CHAPTER 2
The increase of atmospheric methane (CH4) is the second largest contributor to the increased radiative forcing since the industrial revolution. Natural gas extraction and distribution is associated with CH4 leaks of significantly uncertain magnitude that has spurred interest for developing new methods to measure them. Typically, global CH4 emissions related to the oil and gas industry (up-stream, mid-stream and downstream) are estimated at 69-88TgCH4 of the total of 340-360Tg CH4 of anthropogenic CH4 [Saunois et al. 2017]. This chapter is based on the published study by Assan et al., [2017], 1which uses a cavity ring-down spectrometer (CRDS), namely a Picarro G2201i, to evaluate its applicability for two methane identification methods commonly used to better constrain emission estimates from natural gas leaks, a) analysis of 13C and 12C ratios, the two most abundant and stable isotopes of carbon, as well, b) the ethane:methane ratio (C2H6:CH4). Initially, the used G2201i instrument is only specified to measure 12CH4, 13CH4, 12CO2, 13CO2 and H2O by the manufacturer. However, during this work it was found that CRDS measurements of δ13CH4 in the near infrared spectral domain are subject to significant cross sensitivities due to absorption from multiple gases, especially C2H6. The study presents extensive laboratory tests to characterize these cross sensitivities and propose corrections for the biases they induce as well as allow to perform calibrated C2H6 measurements on all G2201i series instruments. Two G2201i instruments were tested to determine the interference of CO2, CH4, and H2O concentrations on C2H6 measurements, and the interference of C2H6 on reported δ13CH4. Methane isotopic measurement were biased to heavier values due to the interference caused by elevated C2H6 concentrations (a secondary component in many natural gas types) by +23.5‰ ppm CH4 /ppm C2H6. The reported C2H6 displays a small sensitivity to absorption interferences from CO2 and CH4, but the predominant interference results from water vapor (with an average linear sensitivity of 0.9 ppm C2H6 per % H2O in ambient conditions, meaning that the presence of H2O causes the inference of too high C2H6 mixing ratios if no correction is applied). Yet, this sensitivity was found to be discontinuous with a strong hysteresis effect. Throughout the range of C2H6 concentrations measured in this study (0-5ppm C2H6), which is large enough to reflect concentrations seen at industrial sites, both CRDS instruments consistently measure concentrations double that reported by a calibrated gas chromatograph, thus we have calculated a calibration factor of 0.5. The generalizability of the corrections and calibrations were determined by repeating the experiments in the study multiple times over the course of a year on two instruments. The study found the calibration factors to be stable in time and between instruments if H2O is kept < 0.16%, to avoid any hysteresis effect. To demonstrate the significance of the corrections, the study tested two source identification methods based on δ13CH4 and C2H6:CH4 of air measured at a natural gas compressor station. The presence of C2H6 in natural gas emissions at an average ambient concentration of 0.3ppm was found to shift the reported isotopic signature by 2.5‰. Furthermore, after correction and calibration the average reported C2H6:CH4 ratio shifts by +0.06. These results indicate that when using such CRDS instruments in conditions of elevated C2H6 for CH4 source determination it is imperative to account for the biases discussed and corrected within this study. Both δ13CH4 and C2H6:CH4 methods were able to correctly distinguish a biogenic source from the on-site natural gas sources; moreover the study found that combining the two independent methods presented a clearer fingerprint of the sources.
With increasing efforts to mitigate anthropogenic greenhouse gas emissions, opportunities to reduce leaks from fossil fuel derived methane (ffCH4) is of particular importance as they currently account for approximately 30% of all anthropogenic methane emissions [Kirschke et al., 2013]. At present, technically feasible mitigation methods hold the potential to half future global anthropogenic CH4 emissions by 2030. Of this mitigation potential more than 60% can be realised in the fossil fuel industry [Hoglund-Isaksson, 2012]. However for effective implementation, sources, locations and magnitudes of emissions must be well known.
The global increase in the production and utilisation of natural gas, of which methane is the primary component, has brought to light questions in regards to its associated fugitive emissions, i.e. leaks. Recent estimates of CH4 leaks vary widely (1-10% of global production) [Allen et al., 2014] and US inventories of natural gas CH4 emissions have uncertainties of up to 30% [EPA, 2016]. To address this issue the ability to distinguish between biogenic and different anthropogenic sources is of vital importance. For this reason methane isotopes (δ13CH4) are commonly used to better understand global and local emissions as demonstrated in a number of studies [Lamb et al., 1995, Lowry et al., 2001, Hiller et al., 2014]. The discrimination of sources with relatively close isotopic composition such as associated-oil gas and natural gas, whose isotopic signatures can be separated by only ~4 ‰ [Stevens et al., 1988], requires precise and reliable δ13CH4 measurements.
Ethane (C2H6) is a secondary component in natural gas and can be used as a marker to distinguish between different CH4 sources. Use of the C2H6:CH4 ratio provides a robust identifier for the gas of interest. Recent findings in the US found coal bed C2H6:CH4 ratios ranging between 0-0.045, while dry and wet gas sources displayed differing ratios of <0.06 and >0.06 respectively [Yacovitch et al., 2014, Roscioli et al., 2015].
Laser spectrometers, especially based on Cavity Ring Down Spectroscopy (CRDS) are now a common deployment for site-scale CH4 measurement campaigns [Yvon-Lewis et al., 2011, Phillips et al., 2013, Subramanian et al., 2015]. However, with the advent of such novel technologies, there lies the risk of unknown interference of laser absorption which can cause biases to measurements. Some examples of which are discussed in Rella et al., (2015) and many others [e.g. K.Malowany et al., 2015, Vogel et al., 2013, Nara et al., 2012]. Using a CRDS instrument we show that the presence of C2H6 is causing significant interference on the measured 13CH4 spectral lines thus resulting in shifted reported δ13CH4 values. We propose a method to correct these interferences, and test it on measurements of natural gas samples performed at an industrial natural gas site.
The CRDS instruments used throughout this study are Picarro G2201-i analysers (Picarro INC, Santa Clara, USA) whose measured gasses include CH4, CO2, H2O, and, although not intended for use by standard users, C2H6. This model measures in 3 spectral ranges; lasers measuring spectral lines at roughly 6057cm-1, 6251cm-1 and 6029cm-1 are used to quantify mole fractions of 12CH4, 12CO2 and 13CO2, and 13CH4, H2O and C2H6 respectively. The spectrograms are fit with two non-linear models in order to determine concentrations; the primary fit is performed excluding the model function of C2H6 while the second includes this function thus adding the ability to measure C2H6 [Rella et al., 2015]. Such a method for measuring C2H6 concentrations is crude, thus the uncalibrated C2H6 concentration data is stored in private archived files which until now have been used primarily for the detection of sample contamination. The measurements of δ13CH4 and δ13CO2 are calculated using the ratios of the concentrations of 12CH4, 13CH4, 12CO2 and 13CO2 respectively.
Table of contents :
Chapter 1 Introduction
1.1 The role of Methane in global warming and climate change
1.2 The Future of Methane
1.3 Uncertainties in Methane emissions estimates
1.4 Anthropogenic Methane Source identification
1.4.1 Methane Formation
1.4.2 VOC Emission Ratios
1.4.3 Stable Isotopes of Methane
1.5 Thesis
1.5.1 Instrumentation
1.5.2 Source Apportionment: developments and tests
1.5.3 Application to field measurements
Chapter 2 Characterisation of interferences to in-situ observations of methane isotopes and C2H6 when using a Cavity Ring Down Spectrometer
2.1 Summary of Chapter 2
2.2 Introduction
2.3 Methods
2.3.1 Experimental Setup
2.3.2 C2H6 calibration setup
2.3.3 Determining the correction for isotopes
2.3.4 Calibration of isotopes
2.4 Results and Discussion
2.4.1 Correcting reported C2H6
2.4.2 C2H6 calibration
2.4.3 Isotopic correction
2.4.4 Isotopic calibration
2.4.5 Typical instrumental performance and uncertainties
2.4.6 Generalisability of corrections and calibrations
2.5 Source Identification at a Natural Gas Compressor Station
2.5.1 Description of field campaign
2.5.2 Impact of C2H6 on isotopic observations at the field site
2.5.3 Continuous field measurements of ethane
2.5.4 Use of continuous observations of C2H6: CH4 by CRDS
2.5.5 Combined method for CH4 source apportionment
2.6 Concluding Remarks
Chapter 3 Can we separate industrial CH4 emission sources from atmospheric observations? – A test case for carbon isotopes, PMF and enhanced APC
3.1 Introduction
3.2 Methods
3.2.1 Description of Dataset
3.2.2 Methods used for source apportionment
3.3 Results & Discussions
3.3.1 Observations at the natural gas compressor station
3.3.2 Analysing the time-series using isotopic data
3.3.3 Analysing the time-series using modified APCA
3.3.4 Analysing the time-series using PMF
3.4 Conclusions
3.5 Supplemental materials: Sensitivity studies
S3.1 Creation of pseudo data
S3.2 Sensitivity Tests
S3.3 Sensitivity Test Results
Chapter 4 Fugitive methane source characterisation of biogenic sources in the Ile de France region.
4.1 Characterising CH4 emissions from dairy farming in Ile-de-France
4.1.1 Site Description of the Grignon farm
4.1.2 Mobile Campaign: 1st May 2017, 9am – Midday
4.1.3. Autumn Field Campaign: 19th Oct – 27th Nov 2016
4.1.4 Spring Field Campaign: 10th of April until the 1st May 2017
4.1.5 Comparison of Autumn & Spring campaigns
4.1.6 Grignon farm Conclusion
4.2. CH4 Emissions from the waste management sector
4.2.1 Waste Water Treatment Facility: St Thibault-des-Vignes
4.2.2 Waste Water Treatment Facility: Cergy Pontoise
4.2.3 Landfill: Butte Bellot
4.2.4 Waste Management Conclusion
4.3 Conclusion
Chapter 5 Thesis Conclusions & Outlooks
Chapter 6 References
